事物发展,既有趋向,又有变异。趋向是事物长期发展的总方向,变异则是发展长途中出现的曲折和反复,是在某一具体时段内偏离长期发展趋势的情况。

This paper has reviewed the classic concept of the cycle of erosion of William Morris Davis (Davis, 1899) on the basis of present rates of uplift and of different land-masses, and analysed the variaties within long1 tendency landscape development both from the viewpoint of unity of opposites and feedback and that of different time and space spans in landscape evolution. Several remarks are summarized as follows:1. The classic model of the cycle of erosion of Davis provided a theoretical framework in landform evolution through time.lt was a great impetus to geomorphological study. However, his view of gradual chang has also some negative influence on the comprehensive understanding of geomorphic processes, and his scheme is too simplistic to interpret the variety of complex natural phenomena.2. Geomorphic cycle is the unity of opposites between variation and feedback (representing tendency). Multiplicity of cycles is characteristic of landform evolution for both variations (or events) and feedbacks occur over and over again in a landscape system. The multiplicity of causes of events results in a number of cycles with different time and space spans. Owing both to the randomness of events and the time lag of feedbacks, the effect of feedbacks in a system will attain different levels. That is why the landscape of a region, is inevitably stamped with the brand of polycycles characteristic of that region. As a result, it is difficult for the relief of any region to be completely levelled and in a landscape system there is no strict equilibrium. That is why the geomorphic types and processes in the world are quite complicated and that there are no totally identical landscapes found in any two regions in the world. On the other hand, the landscape of a region, under the action of the exogenic forces, will tend to be levelled and approach the sea level, and within the system itself there will be coordination between the dynamics and forms through the action of feedbacks. Therefore, there are also numerous similarities among landscapes belonging to the same type in the world.3. Every cycle begins from an event and, undergoes a longer span of feedback time and fmally comes to an end before a new event occurs.After a depositional form is formed, feedback will be going on during a longer time span mainly of self adjustment, including the upbuilding and downcutting of relatively small magnitude. So, in general, geomorphic processes can be either of the four states: (1) the sudden landforming event (variation); (2) the adjustment of the form afterthe event (graded state); (3) the equilibrium state of course with small fluctuations above and below a mean states; and (4) the steady state under which the form is almost constant.4. On studying1 landform evolution, the trammels of the progressive idea must be smashed. We cannot just consider that landform is always progressively being lowered down and continually to be levelled in response to denudation. Instead, we must be conscious of the existence of periods of landmass uplift and lowering of base level. As concerns a graded river, we must be aware that, on the one hand, there is its adaptability to erosion base level and an equilibrium between sediment load and transportability caused by self-regulation and, on the other hand, there are also the stages of form variance due to aggradation or degradation. In respect to a equilibrium stream in which on the average the inflow of sediment equals to the outflow, the local and transitory scour and fill may still exist. In the case that only the long tendency is kept in mind, while neglecting the variation, we shall come to a poor comprehension of the complexity of change of landforms in space and time. On the contrary if we keep out sight only at the cycle interruptions and neglect the long tendency of landform development, well also be misled by transient and local phenomena and cannot utilize geomorphology to its full profit.